Interactive Visualization with R - Interactive Maps - 1
One should look for what is and not what he thinks should be. (Albert Einstein)
Interactive maps: topic introduction
In this part of the course, we will cover the following concepts:
Transform and prepare data for maps
Create maps and display spatial data over time
Warm-Up
Take a few minutes to craft a “tweet” of about 140 characters that either:
poses a question about something we’ve discussed
sums up something valuable you’ve learned
offers a resource you’ve discovered on your own
asks for additional information
When you’re done, post your tweet in the chat so we can pick out a few to discuss in more detail
Module completion checklist
Objective
Complete
Create interactive maps utilizing JSON files
Illustrate adding motion to maps to display spatial data over time
Discuss best practices for highcharter maps
Why interactive maps?
Visualizing spatial data can uncover the interaction between variables and geographic locations
Interactive maps are helpful in visualizing spatial data because of the following:
The ability to create layers of information
Zoom functions and tooltips to show details of a specific point or area
Animations to show the effect of time
Process of creating interactive maps
Set up
Specify data
Link data to visuals
Assign shapes
Adjust
Visual effects
Interactive effects
Legend
Polish
Customize theme
Map layers
Text
Directory settings
In order to maximize the efficiency of your workflow, use the box package and encode your directory structure into variables
Let the main_dir be the variable corresponding to your materials folder
# Set `main_dir` to the location of your materials folder.path = box::file()main_dir =dirname(dirname(path))
Directory settings (cont’d)
We will store all datasets in the data directory inside the materials folder in your environment; hence we will save their path to a data_dir variable
We will save all the plots in the plots directory corresponding to plot_dir variable
To append one string to another, use paste0 command and pass the strings you would like to paste together
# Make `data_dir` from the `main_dir` and# remainder of the path to data directory.data_dir =paste0(main_dir, "/data")# Make `plots_dir` from the `main_dir` and# remainder of the path to plots directory.plot_dir =paste0(main_dir, "/plots")
Set up: state.x77 data
For the non-spatial attributes, we will use a dataset containing facts and figures about U.S. states in 1975, which can be found in R datasets library
For ease of usage, we have stored the dataset in the data folder and we will be loading it directly from the data folder
# We will load the formatted dataset from a csv file.# Load the dataset.state_df =read.csv(file =file.path(data_dir,"/state_data.csv"), #<- provide file pathheader =TRUE, #<- if file has header set to TRUEstringsAsFactors =FALSE) #<- read strings as characters, not as factors
Set up: state.x77 data
Let’s see what variables are represented in the data
You can reference these by link or download and load them using the jsonlite package
# Load the library.library(jsonlite)# View documentation.library(help ="jsonlite")?fromJSON
Set up: working with GEO data and JSON files
We need to give the function a geoJSON file with the spatial data as the first argument
Then we need to tell the function not to simplify any vectors
This setting is necessary for correct object translation into the map data
library(highcharter)
Registered S3 method overwritten by 'quantmod':
method from
as.zoo.data.frame zoo
Highcharts (www.highcharts.com) is a Highsoft software product which is
not free for commercial and Governmental use
# Set working directory to data folder.setwd(data_dir)# Read data from JSON file, don't simplify vectors.US_map =fromJSON("us-all.geo.json", simplifyVector =FALSE)
Set up: working with GEO data and JSON files
The hc_middle_x and hc_middle-y coordinates will be used for plotting
The name and postal-code will be used to create and label the states in the map
# To see what metadata is available in the `geo.json`, use `get_data_from_map` function.geodata =get_data_from_map(US_map)# Look at only 15 first columnsstr(geodata[,1:15])
As always, we first create the empty base, this time of the map type
We use the function hc_add_series to plot the base as the US map
# Create a base interactive map.interactive_population_map =highchart(type ="map") %>%#<- base plot hc_add_series(mapData = US_map)#<- map series
# This is just our base plot.interactive_population_map
Set up: preparing map data
We must join the data in US_map with the data in state_df
This way, the plotting function knows what value must be assigned to what shape
We will use highcharters’ joinBy argument, a vector of 2 variable names, each of which should correspond to variables of the same values in both datasets
In our scenario, let’s use variables that identify the name of the state
These variables are:
The property name in the US_map
The column State in the state_df dataframe
Set up: preparing map data (cont’d)
# Select columns to display on map.data_for_map =select(state_df, Population, #<- select `Population` to display as value on shape State) #<- select `State` to join this data with map data# Rename columns: `Population` -> `value`, because highcharts needs a column # called `value` to attach it to shape.colnames(data_for_map) =c("value", "State") # Adjust data (divide population by 1000, to make units in millions).data_for_map$value = data_for_map$value/1000head(data_for_map)
value State
1 3.615 Alabama
2 0.365 Alaska
3 2.212 Arizona
4 2.110 Arkansas
5 21.198 California
6 2.541 Colorado
Set up: color palettes in maps
Use continuous color palettes to indicate ranges of values from small to large
For example, we can use a range of colors generated based on the state population
The state with the smallest population will be the lightest
The state with the highest population will be the darkest
Color codes (both RGB and hex) can be found on this link
Color palettes can also be created using the color_stops() function
Set up: creating an interactive population map
Let’s combine our input data files and color palette parameters to create an interactive map
We will use the data argument in the hc_add_series function to plot the columns from the cleaned data
We use the function hc_colorAxis to add the color gradient for representing the population in different states
interactive_population_map =highchart(type ="map") %>%hc_add_series(mapData = US_map,data = data_for_map, #<- data to plot on shapesname ="Population in 1975", #<- series name is `Population`joinBy =c("name", #<- join by `name` property in `mapData`"State") ) %>%#<- with `State` column in `data`hc_colorAxis(min =min(data_for_map$value), #<- set colors: min value => minimum populationmax =max(data_for_map$value), #<- max value => maximum populationminColor ="#CCCCFF", #<- min value colormaxColor ="#000066") #<- max value color
Set up: creating an interactive population map
interactive_population_map
Adjust: creating an interactive population map
Let’s add some details to the previous map to improve its functionality for end users
interactive_population_map_adjusted = interactive_population_map %>%hc_add_series(mapData = US_map,data = data_for_map, name ="Population in 1975 ", joinBy =c("name", "State"),dataLabels =list(enabled =TRUE, #<- Add labels with the format =# postal code of the state'{point.properties.postal-code}') ) %>%hc_tooltip(valueSuffix =" million") %>%#<- set value suffixhc_mapNavigation(enabled =TRUE) %>%#<- Add zoom featurehc_title(text ="US States Population in millions (1975)") #<- set plot title
Adjust: creating an interactive population map
# Double click on area or use the `+` button to zoom in on an area.# Use the `-` button to zoom out. interactive_population_map_adjusted
Save interactive plots: htmlwidgets
Finally, let’s save our plot as a self-contained HTML file
# Set working directory to where you save plots.setwd(plot_dir)# Save desired interactive plot to an HTML file.saveWidget(interactive_population_map_adjusted, #<- plot object to save"interactive_population_map.html", #<- name of file to where the plot is to be savedselfcontained =TRUE) #<- set `selfcontained` to TRUE, so that # all necessary files and scripts are embedded # into the HTML file itself
Module completion checklist
Objective
Complete
Create interactive maps utilizing JSON files
✔
Illustrate adding motion to maps to display spatial data over time
Discuss best practices for highcharter maps
Adding motion to maps
Motion in maps can help visualize patterns in data over time and space
# Set working directory to data_dirsetwd(data_dir)# We can also load non-spatial data from a JSON file. data_for_map =fromJSON("drug_overdose.json")head(data_for_map)
fips year value
1 01001 2002 1
2 01003 2002 2
3 01005 2002 0
4 01007 2002 1
5 01009 2002 2
6 01011 2002 0
fips: numeric codes to uniquely identify a county in the U.S.
year: year for which data was collected
value: number of deaths due to drug overdose
Preparing the data
Let’s create a list containing a data series for each year based on three sets of values from the dataset:
fips: we will use this to join with the geoJSON data to create the map
sequence: a list of values in the given state for each year
value: the first value in the list to initialize the map
ds <- data_for_map %>%group_by(fips) %>%#<- group by statedo(state_deaths =list( #<- go through each state fips =first(.$fips), #<- select the FIPS for that statesequence = .$value, #<- create a list of values for that statevalue =first(.$value))) %>%#<- select the first value to initialize the map .$state_deaths #<- put the three together in a list `state_deaths`
Adding motion to maps
The hc_motion function is commonly used to create dynamic and animated charts in highcharter, allowing you to visualize data changes over time or other sequential steps.
Parameters:
enabled (boolean): Enables or disables motion in the plot.
axisLabel (string): Sets the label for the motion slider.
labels (vector): Labels the animation steps (e.g., by year or time period).
Adding motion to maps (cont’d)
Now we can add the interactivity and motion elements to the base map plot
interactive_map_motion <-highchart(type ="map") %>%hc_add_series(data = ds, #<- data to plot on shapesname ="Drug deaths per 100,000", #<- series namemapData = uscountygeojson, #<- map data from `highcharter` joinBy ="fips", #<- join by `fips` field in `data` and `mapData`borderWidth =0.01) %>%#<- adjust border line for each shapehc_colorAxis(stops =color_stops()) %>%#<- create a color legend hc_title(text ="Drug Overdose Deaths per 100,000 between 2002 and 2014") %>%hc_motion( #<- add motionenabled =TRUE,axisLabel ="year", #<- name of motion sliderlabels =sort(unique(data_for_map$year)), #<- label the animation by year )
Visualizing the interactive map
interactive_map_motion
Save interactive plots: htmlwidgets
Remember to save your work!
# Set working directory to where you save plots.setwd(plot_dir)# Save desired interactive plot to an HTML file.saveWidget(interactive_map_motion, #<- plot object to save"interactive_map_motion.html", #<- name of file to where the plot is to be savedselfcontained =TRUE) #<- set `selfcontained` to TRUE, so that # all necessary files and scripts are embedded # into the HTML file itself
Module completion checklist
Objective
Complete
Create interactive maps utilizing JSON files
✔
Illustrate adding motion to maps to display spatial data over time
✔
Discuss best practices for highcharter maps
Tips for map visualization
Aggregate the data correctly for your analysis, i.e., by country, state, or city
Example: Aggregate sales data by summing sales for each city within a state when creating a map showing total sales by state.
Use gradient color scales on maps when plotting continuous or ordinal values
Example: Represent population density by state using a gradient color scale, where higher densities are shown with darker shades and lower densities with lighter shades.
Layer different data on the same map to see variable interactions
Example: Create a map displaying both population density and median income by state to analyze potential patterns or correlations between income and population distribution.
Tips for map visualization (cont’d)
Use motion to add dynamic time-series data to spatial data
Example: Animate a map showing historical temperature data by month for different cities, illustrating temperature patterns throughout the year.
To declutter a map that looks too busy:
eliminate or lighten unnecessary boundaries
use easily interpreted abbreviations wherever possible
use color with some degree of transparency
Example: When creating a map of the United States with state boundaries, decrease the opacity of state boundaries, and use common state abbreviations (e.g., “CA” for California) for improved readability.
